YANMAR Technical Review

Abstract

Considerable progress is being made across a variety of fields in the development of smart technologies based on ICT and IoT with the goal of realizing “A Sustainable Future”. In the case of agriculture, technology for improving the efficiency of agricultural management through the visualization of yield information has been put to practical use and continues to evolve. Yanmar has developed and commercialized a system for combine harvesters that measures yield in real-time. In order to provide this visualization solution to more people, we have developed a yield sensor (YM-KIT, 1150) for use on such harvesters.

Yanmar intends to continue pursuing this work with the aims of further enhancing yield visualization and helping to develop the food-value-chain.

1.Introduction

Considerable progress is being made across a variety of fields on the development of smart technologies based on information and communication technology (ICT) and the Internet of Things (IoT) with the goal of realizing a sustainable future. In the case of agriculture, this has included the development and commercialization of ICT for quantifying agricultural work and crop condition. A key benefit of these technologies is that they can use data to enhance the precision of cultivation management practices which until now had relied solely upon the experience and intuition of the farmer. This has the potential to improve agriculture business performance in the form of increased yields and reduced deployment of resources. As transforming experience into tangible knowledge (data) also facilitates the transfer of farming skills, it can also prevent the loss of expertise resulting from the dramatic fall in the size of the agricultural workforce over recent years. Yanmar believes that by enabling sustainable agriculture, this will contribute to its goal of realizing “A Sustainable Future”.

2.Development Objectives

Yield is among the key candidates for quantification as a means of achieving greater precision in agricultural management. This is because quantifying the yield of each field and presenting this information in visual form can help plan fertilizer use for the following growing season with greater accuracy. Yanmar has developed and commercialized a system for the visualization of yield that works by measuring the quantity of harvested grain as it moves through the combine harvester in real time, and has installed the system to models of head-feeding combine harvesters. Furthermore, Yanmar has also developed a yield sensor (YM-KIT, 1150) that can be fitted to full-feeding combine harvesters, making this yield visualization solution available to an even greater number of agricultural producers. This report describes these new developments.

3.Yield Sensor for Full-feeding Combine Harvesters

3.1.Differences between Grain Conveyors of Head-Feeding and Full-Feeding Combine Harvesters

While the grain conveyors of head-feeding and full-feeding combine harvesters both serve the same purpose of transporting the grain from the bottom of the thresher to the top of the grain tank, they do so in different ways.

Whereas head-feeding combine harvesters use a screw conveyor (see Fig. 1), full-feeding combine harvesters use a bucket conveyor that transports the grain more gently (see Fig. 2). This is because head-feeding combine harvesters are used exclusively for grains such as rice, wheat, or barley that have small kernels and are comparatively resistant to being damaged by external force. Full-feeding combine harvesters, in contrast, are also used to harvest crops such as soy beans that have large kernels and are more easily damaged.

As with head-feeding combines, the yield in full-feeding combines is measured as the grain travels from the thresher to the grain tank, by a yield sensor located at the discharge point of the bucket conveyor (see Fig. 3). However, due to differences in the grain transport method described above, modifications are needed in terms of the exact location of the yield sensor, its design, and the method by which it calculates the yield.

3.2.Location of Yield Sensor on Full-Feeding Combine Harvesters

The bucket conveyor consists of deep bucket-like containers, each of which is able to hold a large amount of grain. Accordingly, not all of the grain contained within a bucket is discharged simultaneously when the bucket begins to tip as it reaches the top of the conveyor. Moreover, the grain is discharged from the buckets in two stages. Some of the grain is ejected in the radial direction due to the centrifugal force as the bucket travels around the axis of the conveyor (mechanism A), while the rest is ejected by inertia as the bucket changes from circular to linear motion (mechanism B) (see Fig. 4). While all of the grain left behind after mechanism A is ejected by mechanism B, a major feature of bucket discharge is that the relative proportions differ depending on crop conditions.

3.3.Yield Sensor for Full-Feeding Combine Harvesters

The yield sensor for full-feeding combine harvesters calculates the amount of grain by using the same small load cell as used on head-feeding combine harvesters to measure the force of impact of discharged grain as it hits the sensor, and converting this into an electrical signal. Unfortunately, because the mechanism used by full-feeding combine harvesters to transport the grain to the grain tank is different to that of head-feeding combine harvesters (as explained above), it is not practical simply to replicate the same sensor position, design, and yield calculation as used on head-feeding combine harvesters. In addition, with full-feeding combines harvesting a variety of crops, the different grain discharge ratios associated with different crops/crop conditions present significant challenges in obtaining accurate yield measurements.

In order to overcome these challenges, the following two improvements were made to the yield measurement in full-feeding combine harvesters: [1] A guide was installed, making it possible to measure more of the radially discharged grains, and [2] Sensors were installed in two locations. Grains discharged radially and collected by the guide are measured by sensor A. Grains expelled downwards are measured by sensor B. This method enables as many grains as possible to be measured, and provides highly accurate results regardless of variations in crop condition (see Fig. 5).

A preliminary study was conducted using simulation to visualize the grain flow and aid in the analysis. This provided information on how the sensor output is affected by the sensor position, the guide shape, and the grain properties.

During harvesting, variations in field conditions (e.g. crop variety and moisture content) result in changes in the physical properties of grains, such as their friction coefficient and rolling coefficient. This section describes some of the work done to study how these changing physical properties affect the yield sensor output. The study found that changes in the friction coefficient and rolling coefficient have a large influence on the outputs of sensors A and B. This indicated that the variability in the physical properties of the grain would make it difficult to achieve adequate accuracy using the measurements from a single location (sensor A or B). The work also demonstrated that, as a means of dealing with this variability, the degree of measurement disparity could be reduced by installing two sensors and adding their results together (see Fig. 6).

3.4.Harvest Rate and Cumulative Yield in Real Time

Field trials were conducted to verify the yield measurement accuracy. The system is currently able to perform real-time measurement/visualization of the rate of harvest and cumulative yield for rice and wheat (see Fig. 7). By linking the system to Yanmar’s SmartAssist, a solution that provides producers with total support, and by using it to manage and utilize the data, Yanmar intends to support operational improvements and help make crop planning more efficient (see Fig. 8).

4.Conclusions

The use of a yield sensor for combine harvesters to enable the visualization of crop yields represents another small step toward a world in which operational analysis, planning, and management are easily performed. Yanmar is pursuing exciting developments with a view to the potential uses of data able to be collected during farm operations and its transformation into valuable information through integration with other systems. Yanmar also intends to continue contributing to the food and agricultural industries by further pursuing the possibilities of visualization.

-IMPORTANT-

The original technical report is written in Japanese.

This document was translated by Research & Development Management Division.